Page Banner

United States Department of Agriculture

Agricultural Research Service


Location: Hydrology and Remote Sensing Laboratory

Title: PP-SWAT: A phython-based computing software for efficient multiobjective callibration of SWAT

item Zhang, Xuesong
item Beeson, Peter
item Link, Robert
item Manowitz, David
item Sadeghi, Ali
item Thomson, Allison
item Shahajapal, R
item Sirinivasan, R
item Arnold, Jeffrey

Submitted to: Journal of Environmental Modeling and Software
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/13/2013
Publication Date: 4/28/2013
Publication URL:
Citation: Zhang, X., Beeson, P.C., Link, R., Manowitz, D., Sadeghi, A.M., Thomson, A.M., Shahajapal, R., Sirinivasan, R., Arnold, J.G. 2013. Efficient multi-objective callibration of a computationally intensive hydrological model with parallel computing software in python. Journal of Environmental Modeling and Software. 46:208-218.

Interpretive Summary: In this investigation, we have developed a new computer software, called “Python-based Parallel Soil and Water Assessment Tool (or PP-SWAT),” for efficient calibration of SWAT, a USDA watershed model. This software allows for simultaneously optimizing streamflow simulations at multiple sites (as opposed to a single site calibration) and can support additional key hydrologic processes, such as tile flow, within a watershed. We successfully tested PP-SWAT in two watersheds (Little River Experimental Watershed in Georgia and the South Fork watershed in Iowa) on the University of Maryland's Evergreen supercomputer. Results showed that PP-SWAT allowed finer scale parameter adjustments, helped modelers to have a much wider range of options of parameters in defining parameter values, and to achieve significant time savings for long simulations (over 50 times faster). Overall, PP-SWAT developed here is expected to serve as an efficient tool for calibrating SWAT and diversify the existing base of SWAT automatic calibration software.

Technical Abstract: With enhanced data availability, distributed watershed models for large areas with high spatial and temporal resolution are increasingly used to understand water budgets and examine effects of human activities and climate change/variability on water resources. Developing parallel computing software to improve calibration efficiency has received growing attention of the watershed modeling community as it is very time demanding to iteratively run complex models for calibration. In this research, we introduce a Python-based parallel computing software, PP-SWAT, for efficient calibration of the SWAT model. This software employs Python 2.7, MPI 4py and OpenMPI to parallelize A Multi-method Genetically Adaptively Multi-objective Optimization Algorithm (AMALGAM), allowing for simultaneously addressing multiple objectives in optimizing SWAT. Test results on a Linux computer cluster show that PP-SWAT can achieve a speedup of 45-109 depending on model complexity. It is worth noting that increasing number of processors does not necessarily help improve efficiency because resource competition may slow I/O operations. In addition to achieve improved efficiency, one benefit of PP-SWAT is to allow adjusting parameters at different scales (watershed, grouped-subbasin, or subbasin). In comparison to only adjusting parameters at watershed scale, this flexibility provides SWAT users with a wider range of options of parameter sets to choose from for model(s) selection. We also examined potential effects of model structure initialization and parameter constraints on the performance of PP-SWAT. Results show that even with incorrect model structure and very distinct parameter ranges, PP-SWAT can arrive at satisfactory simulation of streamflow at multi-sites. In this context, expert knowledge and information from other sources on key hydrological processes, such as tile flow, are critical in differentiating parameters. It is suggested cautious supervision of PP-SWAT’s power be exercised in order to produce calibrated models that well represent watershed behavior.

Last Modified: 10/18/2017
Footer Content Back to Top of Page